On Aug. 5, the Federal Circuit reaffirmed its 2-1 prior decision that Teva’s sales of carvedilol, a cardiovascular drug, induced infringement of GSK’s method of treatment patent, even though Teva had filed a section viii statement to carve out the infringing uses from its product label. The majority called its decision a “narrow, case-specific review” that would not affect other cases with section viii carve out, but its impact on the industry has yet to be seen and may in fact be far-reaching.

The court reinstated the $235 million patent infringement jury verdict against Teva, again finding that Teva encouraged doctors to prescribe the generic carvedilol for infringing uses that it had carved out in its Abbreviated New Drug Application (ANDA). The majority opinion, authored by Chief Judge Moore and Judge Newman, found that there was substantial evidence to support the jury verdict during the period Teva was marketing its product with a skinny label because, the majority found, the evidence showed that Teva’s attempted carve-out was ineffective. Specifically, the majority found that one of the indications Teva left in its label, to reduce cardiovascular mortality in patients suffering from left ventricular dysfunction following myocardial infarction (“the post-MI LVD indication”), promoted (along with other portions of the label) the method of decreasing mortality caused by congestive heart failure claimed in GSK’s patent (“CHF patent”). Former Chief Judge Prost dissented, finding that GSK’s CHF patent could not cover the post-MI LVD indication because GSK never identified that indication to FDA as being covered by its CHF patent when it submitted that patent to the Food and Drug Administration (FDA), in accordance with the Hatch-Waxman Act and Federal regulations implementing the Act. For the full-label period (the period Teva marketed its product with the indication covered by GSK’s CHF patent), Judge Prost also dissented from the majority’s opinion finding substantial evidence supporting the jury’s infringement verdict because, in Judge Prost’s opinion, there was no evidence in the record supporting the jury’s finding that Teva’s actions actually caused physicians to prescribe Teva’s product in an infringing manner.

What the Ruling Means

Section 505(j)(2)(A)(viii) of the Food, Drug, and Cosmetic Act, which is also known as the “section viii carve out” or a “skinny label” protects generic drug companies from infringement lawsuits as long as the uses described in the ANDA labeling are not are not covered by the brand company’s patents.

If the relevant patent, typically a method of treatment patent, is listed in the Orange Book only for a drug’s use, the generic company may submit a section viii carve out stating that its ANDA labeling does not claim the same condition of use. The generic company also carves out the protected use information, typically indications or data, from its prescribing information. FDA typically accepts the section viii carve out so long as the skinny label does not make the generic drug less safe or effective.

The October 2020 Federal Circuit opinion in the same case seemed to indicate that a generic company could be held liable for induced infringement by simply selling a generic version of the drug with a skinny label.

In the present decision, published after the rehearing of the October 2020 decision, the Federal Circuit clarified that a straight use of a skinny label by itself could not justify liability for inducement infringement and that an “AB rated” generic, which means interchangeable with the branded product, would not be automatic evidence of inducing infringement. The majority found sufficient evidence that Teva went further – it labeled and marketed the drug in a way that encouraged the patented use (by not carving out the post-MI LVD indication). Though the majority opinion on Teva’s rehearing petition gives generic companies a glimmer of hope that the skinny label is not completely dead, the ruling makes it clear that the skinny label under section viii does not provide an automatic safe-harbor for the generic product.

How the Ruling May Impact Your Business

A $235 million verdict is an eye popping number that will force all generic companies to be more careful on launching at-risk with a skinny label. The re-instatement of this jury verdict by Federal Circuit has reignited debate on section viii carve-outs that was considered settled. For example, the Federal Circuit in AstraZeneca Pharmaceuticals LP v. Apotex Corp., 669 F.3d 1370 (Fed. Cir. 2012) had held that “an ANDA seeking to market a drug not covered by a composition patent for unpatented methods of treatment cannot infringe under [35 U.S.C.] § 271(e)(2).”  Id. at 1379. However, the present decision sends a clear signal that the skinny labels must be carried out with caution as it is no longer provides a safe-harbor that generic companies once thought it did. The opinion makes clear that a drug label, if not carefully written to carve out the infringing use, may still be read as a user manual to support a cause of action for induced infringement.

As a result of this decision, some generic companies may not file ANDA for some of the older drugs whose molecule patent term may have ended but may still be protected by method of treatment patents. Such a setback in the industry is simply not what Congress intended, when it legislated the Hatch Waxman act “to speed the introduction of low-cost generic drugs to market.” Former California Rep. Henry Waxman, the co-author of the 1984 Hatch-Waxman Act, himself filed an amicus brief after the October 2020 decision and argued that decision was inconsistent with the act and could have a “devastating impact on the Hatch-Waxman Act’s generic drug program.”

Procedural History

The carvedilol compound is a beta blocker and was patented in 1985 as U.S. Patent No. 4,503,067 with an expiration date of March 5, 2007. By 1995 GSK had received FDA approval to use carvedilol to treat hypertension. In 1998, GSK was granted U.S. Patent No. 5,760,069, which claimed a method of administering a combination of carvedilol and one or more of an ACE inhibitor, to decrease mortality caused by congestive heart failure in a patient. GSK filed a supplemental NDA to add the congestive heart failure as an indication to the label and listed the ’069 patent to the Orange Book shortly after its issuance. In March 2002, Teva filed an ANDA for FDA approval of its generic carvedilol for both indications that carvedilol was approved for at the time. It filed a paragraph III statement as to the ’067 compound patent and argued that the ’069 method patent was invalid. GSK did not sue or seek to block Teva’s approval immediately. It however sought a reissue of the ’069 patent by narrowing limitation to overcome validity challenges. In 2003, GSK got approval for third indication – post-MI LVD indication. Teva updated its proposed label to include all three indications. In 2004, the FDA tentative approved Teva’s ANDA, such that Teva would be permitted to market its ANDA product in 2007, after the expiration of the ’067 patent. In January 2008, the PTO issued reissue patent – RE40,000 and GSK listed the ’000 patent in the Orange Book for Coreg^®. Just before Teva launched its generic carvedilol in 2007, it carved out under section viii the indication decreasing mortality caused by congestive heart failure, i.e., the one covered by the ’000 patent (reexamined CHF patent).

In 2011, GSK delisted certain patents including the CHF patent but maintained the ’000 patent. Teva amended its label but argued that it need not provide certification to the ’000 patent because its ANDA was approved before the ’000 patent issued. In 2013, GSK sued Teva and Glenmark in Delaware. The case against Glenmark was stayed. After a seven-day jury trial, the jury found Teva induced infringement of the ’000 patent and that its infringement was willful. The district court however granted judgment as a matter of law to Teva stating that the evidence did not support the verdict of induced infringement. Chief Judge Stark noted that there was some overlap between the congestive heart failure and left ventricular dysfunction post myocardial infarction but not sufficient to support the jury verdict. GSK appealed and the Federal Circuit agreed with GSK. The Federal Circuit in its decision, pointed to Teva’s marketing material as touting its generic carvedilol as “indicated for treatment of heart failure and hypertension,” as the “Generic version of [GSK’s] cardiovascular agent Coreg®,” and as an “AB-rated generic equivalent of [GSK’s] Coreg® Tablets.”

Teva requested a rehearing and numerous entities submitted amicus briefs raising concerns because the October 2020 decision could be broadly read to impose liability on the ANDA filers that carve out patented uses under section viii and could be read to upset the careful balance struck in prior Federal Circuit decisions.

The majority clarified that a drug that claims to be an AB-rated generic wouldn’t be automatic evidence of inducing infringement, which was a concern raised by amici and Teva.² The majority focused more on the label than the marketing material in the rehearing opinion, which is sure to raise concerns for the generics. 

Judge Prost dissented in the rehearing opinion again, raising concerns that the decision creates confusion for generics, leaving them in the dark about what might expose them to liability in inducement cases, weakens the standard for inducement in pharmaceutical cases, and throws a wrench into Congress’s design for enabling quick, public access to generic drugs.

The Federal Circuit also denied Teva’s motion for en banc hearing, which means that Teva may now appeal that decision via a cert petition to the Supreme Court.

---

¹ GlaxoSmithKline LLC v. Teva Pharms. USA, Inc., No. 2018-1976, 2021 WL 3412496 (Fed. Cir. Aug. 5, 2021)

² Our partner, Mr. Andrew Alul from Taft Law, had filed an amicus brief on behalf of Apotex Inc. raising this concern and that the evidence the majority relied upon in its first opinion to find substantial evidence of inducement was insufficient under the Federal Circuit’s inducement precedent.